The broad objective of this research program is to understand the relationship between the molecular structure of a polypeptide chain and its ability to fold into a defined, three-dimensional structure. Most studies on protein folding and stability have focused on the role of amino acid side-chains using site-directed mutagenesis. We propose to diverge from this trend by using the total synthesis of proteins to chemically modify the polypeptide backbone. We believe that systematic variation of the backbone will give insight into the fundamental forces that stabilize proteins and the processes through which they fold. We have demonstrated that alpha-hydroxy acids can be incorporated into proteins in a site-specific manner using peptide synthesis and chemical ligation methods. We have utilized this modification to analyze the energetic contributions of specific hydrogen bonds in the GCN4 coiled coil and, in kinetic studies, the folding transition state of the chymotrypsin inhibitor CI2. These studies indicate that backbone modification provides direct information on the formation of backbone hydrogen bonding in the native state and folding transition state ensemble that is not observed using traditional side chain mutagenesis methods. This proposal aims to answer specific questions regarding the folding transition states of GCN4 and CI2 and to extend these studies to the B1 domain of Protein L and acylphosphatase. These proteins have been selected to take advantage of previous work using site directed mutagenesis to analyze the folding transitions of these proteins. In addition, recent computational analyses of these proteins have made specific predictions about folding that cannot be addressed by traditional experimental mutagenesis strategies. We feel that this use of non-coded modifications such as ester bonds for thermodynamic and kinetic measurements of protein folding will enable new insights into the molecular basis of protein folding and stability and provide data for the continued development of computational approaches to this problem.